OpenAI Gym is a open-source Python toolkit for developing and comparing reinforcement learning algorithms. This R package is a wrapper for the OpenAI Gym API, and enables access to an ever-growing variety of environments.
You can install:
the latest released version from CRAN:
install.packages("gym")
the latest development version from Github:
if (packageVersion("devtools") < 1.6) {
install.packages("devtools")
}
devtools::install_github("paulhendricks/gym-R", subdir = "gym")
If you encounter a clear bug, please file a minimal reproducible example on github.
library(gym)
remote_base <- "http://127.0.0.1:5000"
client <- create_GymClient(remote_base)
print(client)
# Create environment
env_id <- "CartPole-v0"
instance_id <- env_create(client, env_id)
print(instance_id)
# List all environments
all_envs <- env_list_all(client)
print(all_envs)
# Set up agent
action_space_info <- env_action_space_info(client, instance_id)
print(action_space_info)
agent <- random_discrete_agent(action_space_info[["n"]])
# Run experiment, with monitor
outdir <- "/tmp/random-agent-results"
env_monitor_start(client, instance_id, outdir, force = TRUE, resume = FALSE)
episode_count <- 100
max_steps <- 200
reward <- 0
done <- FALSE
for (i in 1:episode_count) {
ob <- env_reset(client, instance_id)
for (i in 1:max_steps) {
action <- env_action_space_sample(client, instance_id)
results <- env_step(client, instance_id, action, render = TRUE)
if (results[["done"]]) break
}
}
# Dump result info to disk
env_monitor_close(client, instance_id)
The original author of gym
is Paul Hendricks.
The lead maintainer of gym
is Paul Hendricks.